Convolutional Neural Networks¶
Project: Write an Algorithm for a Dog Identification App¶
In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!
Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.
In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.
The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.
Why We're Here¶
In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).
In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!
The Road Ahead¶
We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.
- Step 0: Import Datasets
- Step 1: Detect Humans
- Step 2: Detect Dogs
- Step 3: Create a CNN to Classify Dog Breeds (from Scratch)
- Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)
- Step 5: Write your Algorithm
- Step 6: Test Your Algorithm
Step 0: Import Datasets¶
Make sure that you've downloaded the required human and dog datasets:
Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location
/dogImages.Download the human dataset. Unzip the folder and place it in the home directory, at location
/lfw.
Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.
In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.
!pip install --upgrade pip
!pip install torch
!pip install torchvision
!pip install -U numpy
!pip show numpy
# Source: https://machinelearningmastery.com/how-to-perform-face-detection-with-classical-and-deep-learning-methods-in-python-with-keras/
!pip install mtcnn==0.0.9
!pip show mtcnn
!pip install --upgrade pillow
import numpy as np
from glob import glob
import cv2
from tqdm import tqdm
import matplotlib.pyplot as plt
from matplotlib.patches import Rectangle
from matplotlib.pyplot import imshow
%matplotlib inline
from keras.utils import np_utils
import torch
import torchvision.models as models
import torchvision.transforms as transforms
from torch.autograd import Variable
from torchvision import datasets, transforms
import io
import requests
from PIL import Image
import os
# the following import is required for training to be robust to truncated images
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
# check if CUDA is available
use_cuda = torch.cuda.is_available()
# load filenames for human and dog images
human_files = np.array(glob("/data/lfw/*/*"))
dog_files = np.array(glob("/data/dog_images/*/*/*"))
# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
Step 1: Detect Humans¶
In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.
OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.
# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')
# load color (BGR) image
img = cv2.imread(human_files[100])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# find faces in image
faces = face_cascade.detectMultiScale(gray)
# print number of faces detected in the image
print('Number of faces detected:', len(faces))
# get bounding box for each detected face
for (x,y,w,h) in faces:
# add bounding box to color image
cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.
In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.
Write a Human Face Detector¶
We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
img = cv2.imread(img_path)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray)
return len(faces) > 0
(IMPLEMENTATION) Assess the Human Face Detector¶
Question 1: Use the code cell below to test the performance of the face_detector function.
- What percentage of the first 100 images in
human_fileshave a detected human face? - What percentage of the first 100 images in
dog_fileshave a detected human face?
Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.
Answer: (You can print out your results and/or write your percentages in this cell)
human_files_short = human_files[:100]
dog_files_short = dog_files[:100]
#-#-# Do NOT modify the code above this line. #-#-#
## TODO: Test the performance of the face_detector algorithm
## on the images in human_files_short and dog_files_short.
# counter for detecting human faces in human_files_short
count_human_faces = 0
# counter for detecting dog faces as human faces in dog_files_short
count_dog_faces_as_human_faces = 0
for image_path_human, image_path_dog in zip(human_files_short, dog_files_short):
if face_detector(image_path_human) == True:
count_human_faces += 1
if face_detector(image_path_dog) == True:
count_dog_faces_as_human_faces += 1
print('Percentage of detected human faces: ', count_human_faces, '%')
print('Percentage of detected dog faces as human faces: ', count_dog_faces_as_human_faces,'%')
Face Detection With Deep Learning¶
A number of deep learning methods have been developed and demonstrated for face detection.
Perhaps one of the more popular approaches is called the “Multi-Task Cascaded Convolutional Neural Network,” or MTCNN for short, described by Kaipeng Zhang, et al. in the 2016 paper titled “Joint Face Detection and Alignment Using Multitask Cascaded Convolutional Networks.”
The following code lines are inspired by How to Perform Face Detection with Deep Learning in Keras
Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
### (Optional)
### TODO: Test performance of anotherface detection algorithm.
### Feel free to use as many code cells as needed.
# face detection with mtcnn on a photograph
from mtcnn.mtcnn import MTCNN
# create the detector, using default weights
detector = MTCNN()
human_files_short = human_files[:100]
dog_files_short = dog_files[:100]
# draw an image with detected objects
def draw_image_with_boxes(filename, result_list, index):
# load the image
data = plt.imread(filename)
# plot the image
plt.imshow(data)
# get the context for drawing boxes
ax = plt.gca()
# plot each box
for result in result_list:
# get coordinates
x, y, width, height = result['box']
# create the shape
rect = Rectangle((x, y), width, height, fill=False, color='red')
# draw the box
ax.add_patch(rect)
# show the plot
plt.show()
# counter for detecting human faces in human_files_short
count_human_faces = 0
# counter for detecting dog faces as human faces in dog_files_short
count_dog_faces_as_human_faces = 0
index = 0
for image_path_human, image_path_dog in zip(human_files_short, dog_files_short):
#pixels_human = plt.imread(image_path_human)
#pixels_dog = plt.imread(image_path_dog)
pixels_human = cv2.cvtColor(cv2.imread(image_path_human), cv2.COLOR_BGR2RGB)
pixels_dog = cv2.cvtColor(cv2.imread(image_path_dog), cv2.COLOR_BGR2RGB)
# detect faces in the image
faces_human = detector.detect_faces(pixels_human)
faces_dog = detector.detect_faces(pixels_dog)
# display faces on the original image
if index < 20:
draw_image_with_boxes(image_path_human, faces_human, index)
draw_image_with_boxes(image_path_dog, faces_dog, index)
index += 1
for face in faces_human:
if len(face['box']) >0:
count_human_faces += 1
break
for face in faces_dog:
if len(face['box']) >0:
count_dog_faces_as_human_faces += 1
break
print('Percentage of detected human faces: ', count_human_faces, '%')
print('Percentage of detected dog faces as human faces: ', count_dog_faces_as_human_faces,'%')
Result¶
The usage of the MTCNN face detector trained on a Cascaded Convolutional Neural Network leads to better result with regard to the detection of human faces than the usage of the OpenCV face_detector. 100% of human faces in the human_files_short list were correctly detected. The wrong detection of dog faces in the range of 20% is comparable to the result by using the OpenCV approach, however.
Percentage of detected human faces: 100 % Percentage of detected dog faces as human faces: 16 %
Step 2: Detect Dogs¶
In this section, we use a pre-trained model to detect dogs in images.
Obtain Pre-trained VGG-16 Model¶
The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.
# define VGG16 model
VGG16 = models.vgg16(pretrained=True)
# check if CUDA is available
use_cuda = torch.cuda.is_available()
# move model to GPU if CUDA is available
if use_cuda:
VGG16 = VGG16.cuda()
print(VGG16)
Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.
(IMPLEMENTATION) Making Predictions with a Pre-trained Model¶
In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.
Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.
# Do necessary transformations to get the right image tensors
def transform_to_tensor(img_path):
# code from Style transfer section
# define transforms for the training data and testing data
transform_img_size = 224
transform_crop = 224
transform = transforms.Compose([transforms.Resize(transform_img_size),
transforms.CenterCrop(transform_crop),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406],
[0.229, 0.224, 0.225])])
# load the image
image = Image.open(img_path).convert('RGB') # Read bytes and store as an img.
image = transform(image)[:3,:,:]
# PyTorch pretrained models expect the Tensor dims to be (num input imgs, num color channels, height, width).
# Currently however, we have (num color channels, height, width); let's fix this by inserting a new axis.
image = image.unsqueeze(0) # Insert the new axis at index 0 i.e. in front of the other axes/dims.
return image
# code from Style transfer section
# and converting it from a Tensor image to a NumPy image for display
def im_convert(tensor):
""" Display a tensor as an image. """
image = tensor.to("cpu").clone().detach()
image = image.numpy().squeeze()
image = image.transpose(1,2,0)
image = image * np.array((0.229, 0.224, 0.225)) + np.array((0.485, 0.456, 0.406))
image = image.clip(0, 1)
return image
# Prediction function based on VGG16
def VGG16_predict(img_path):
'''
Use pre-trained VGG-16 model to obtain index corresponding to
predicted ImageNet class for image at specified path
Args:
img_path: path to an image
Returns:
Index corresponding to VGG-16 model's prediction
'''
## TODO: Complete the function.
## Load and pre-process an image from the given img_path
## Return the *index* of the predicted class for that image
VGG16.eval()
# Implement image transformations
img = transform_to_tensor(img_path)
# get sample outputs
output = VGG16(img) # Returns a Tensor of shape (batch, num class labels)
_, preds_tensor = torch.max(output, 1) # Our prediction will be the index of the class label with the largest value.
#print(labels[prediction]) # Converts the index to a string using our labels dict
return preds_tensor # predicted class index
# Plot the dog image from dog_files_short together with the predicted class (0...999) based on VGG16
for image_path_human, image_path_dog in zip(human_files_short, dog_files_short):
img_path = image_path_dog
# Let's take a look!
img_original = plt.imread(img_path)
# plot the original image
plt.imshow(img_original)
plt.show()
# plot the Tensor image
img = transform_to_tensor(img_path)
plt.imshow(im_convert(img))
plt.show()
#VGG16_predict(dog_files[1])
print('Predicted index: ', VGG16_predict(image_path_dog))
(IMPLEMENTATION) Write a Dog Detector¶
While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).
Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
## TODO: Complete the function.
pred_index = VGG16_predict(img_path)
return ((pred_index <= 268) & (pred_index >= 151)) # true/false
(IMPLEMENTATION) Assess the Dog Detector¶
Question 2: Use the code cell below to test the performance of your dog_detector function.
- What percentage of the images in
human_files_shorthave a detected dog? - What percentage of the images in
dog_files_shorthave a detected dog?
Answer:
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.
counter_human_as_dog = 0
counter_dog_as_dog = 0
for image_path_human, image_path_dog in zip(human_files_short, dog_files_short):
if dog_detector(image_path_human) == True:
counter_human_as_dog += 1
# plot the Tensor image
img = transform_to_tensor(image_path_human)
plt.imshow(im_convert(img))
plt.show()
# VGG16_predict(dog_files[1])
print(VGG16_predict(img_path))
print(dog_detector(img_path))
if dog_detector(image_path_dog) == True:
counter_dog_as_dog += 1
# plot the Tensor image
img = transform_to_tensor(image_path_dog)
plt.imshow(im_convert(img))
plt.show()
# VGG16_predict(dog_files[1])
print(VGG16_predict(img_path))
print(dog_detector(img_path))
print('')
print('Result:')
print('Percentage of human beings detected as dogs: ', counter_human_as_dog, '%')
print('Percentage of dogs detected as dogs: ', counter_dog_as_dog,'%')
We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
### (Optional)
### TODO: Report the performance of another pre-trained network.
### Feel free to use as many code cells as needed.
Step 3: Create a CNN to Classify Dog Breeds (from Scratch)¶
Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.
We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.
| Brittany | Welsh Springer Spaniel |
|---|---|
 |
 |
It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).
| Curly-Coated Retriever | American Water Spaniel |
|---|---|
 |
 |
Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.
| Yellow Labrador | Chocolate Labrador | Black Labrador |
|---|---|---|
 |
 |
 |
We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.
Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!
(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset¶
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!
### TODO: Write data loaders for training, validation, and test sets
## Specify appropriate transforms, and batch_sizes
# number of subprocesses to use for data loading
# code from Data Augmentation Section
data_dir = '/data/dog_images'
# TODO: Define transforms for the training data and testing data
resizing = 224
cropping = 224
train_transforms = transforms.Compose([transforms.RandomRotation(10),
transforms.RandomResizedCrop(cropping),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])])
test_transforms = transforms.Compose([transforms.Resize(resizing),
transforms.CenterCrop(cropping),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])])
# Pass transforms in here, then run the next cell to see how the transforms look
train_data = datasets.ImageFolder(data_dir + '/train', transform=train_transforms)
valid_data = datasets.ImageFolder(data_dir + '/valid', transform=test_transforms)
test_data = datasets.ImageFolder(data_dir + '/test', transform=test_transforms)
train_loader = torch.utils.data.DataLoader(train_data, batch_size=32, shuffle=True)
test_loader = torch.utils.data.DataLoader(test_data, batch_size=16)
valid_loader = torch.utils.data.DataLoader(valid_data, batch_size=16)
loaders_scratch = {'train': train_loader, 'valid': valid_loader, 'test': test_loader}
# Plot images with labels of one batch
%matplotlib inline
# obtain one batch of training images
dataiter = iter(train_loader)
images, labels = dataiter.next()
print('These are the labels from train data:')
print(labels)
print('\n')
print('This is the train_label to index dictionary:')
print(train_data.class_to_idx)
# plot the images in the batch, along with the corresponding labels
fig = plt.figure(figsize=(25, 4))
for idx in np.arange(20):
ax = fig.add_subplot(2, 20/2, idx+1, xticks=[], yticks=[])
ax.imshow(im_convert(images[idx]))
# print out the correct label for each image
# .item() gets the value contained in a Tensor
ax.set_title(str(labels[idx].item()))
Question 3: Describe your chosen procedure for preprocessing the data.
- How does your code resize the images (by cropping, stretching, etc)? What size did you pick for the input tensor, and why?
- Did you decide to augment the dataset? If so, how (through translations, flips, rotations, etc)? If not, why not?
Answer: My answer is based on research under https://pytorch.org/docs/stable/torchvision/transforms.html
For Training I use Data Augmentation with the following transformations:
- transforms.RandomRotation(10) --> random Rotation of the image. The range of degrees will be (-10, +10).
- transforms.RandomResizedCrop(224) --> This will extract a patch of size (224, 224) from the input image randomly.
- transforms.RandomHorizontalFlip() --> Once the image is of size (224, 224), horizontal flipping can be applied.
- transforms.ToTensor() --> This transforms the image to a tensor.
- transforms.Normalize[0.485, 0.456, 0.406], [0.229, 0.224, 0.225]) --> data scaling to normalize the Tensor (1st list = mean, 2nd list = std),values are given by torchvision.
For Validation and Testing I do not use Data Augmentation. The following transformations are aplied:
- transforms.Resize(224) --> smaller edge of the image will be matched to this 224. i.e, if height > width, then image will be rescaled to (size * height / width, size).
- transforms.CenterCrop(224) --> Crops the given image at the center. A square crop (224, 224) is made.
- transforms.ToTensor() --> This transforms the image to a tensor.
- transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])]) --> data scaling to normalize the Tensor (1st list = mean, 2nd list = std),values are given by torchvision.
Based on the documentation of torchvision.models https://pytorch.org/docs/stable/torchvision/models.html I have chosen 224 pixel as the input size. All pre-trained torchvision models expect input images normalized in the same way, i.e. mini-batches of 3-channel RGB images of shape (3 x H x W), where H and W are expected to be at least 224. The images have to be loaded in to a range of [0, 1] and then normalized using mean = [0.485, 0.456, 0.406] and std = [0.229, 0.224, 0.225].
(IMPLEMENTATION) Model Architecture¶
Create a CNN to classify dog breed. Use the template in the code cell below.
import torch.nn as nn
import torch.nn.functional as F
# define the CNN architecture
class Net(nn.Module):
### TODO: choose an architecture, and complete the class
def __init__(self):
super(Net, self).__init__()
## Define layers of a CNN
self.conv1 = nn.Conv2d(3, 16, 3, padding=1)
self.conv2 = nn.Conv2d(16, 32, 3, padding=1)
self.conv3 = nn.Conv2d(32, 64, 3, padding=1)
# max pooling layer
self.pool = nn.MaxPool2d(2, 2)
# dropout (p=0.2)
self.dropout = nn.Dropout(0.2)
# FC Network, classifier
self.fc1 = nn.Linear(28*28*64,500)
self.fc2 = nn.Linear(500,133)
def forward(self, x):
## Define layers of a CNN
x = self.pool(F.relu(self.conv1(x)))
x = self.pool(F.relu(self.conv2(x)))
x = self.pool(F.relu(self.conv3(x)))
x = x.view(-1,28*28*64)
x = self.dropout(x)
x = F.relu(self.fc1(x))
x = self.dropout(x)
x = self.fc2(x)
return x
#-#-# You so NOT have to modify the code below this line. #-#-#
# instantiate the CNN
model_scratch = Net()
# move tensors to GPU if CUDA is available
if use_cuda:
model_scratch.cuda()
print(model_scratch)
Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step.
Answer: The Net class is divided into two functions: The init-function and the forward-function.
In the init-function I define the Conv, MaxPool and FC layers as well as a dropout regularization. There are three Conv-layers for a feature extraction with growing depth from depth=3 via depth=16 to depth=32. The Kernel size of these layers is 3 (3x3 squared window) with a stride and padding of 1. These stride and padding values ensure that the image size (width and height) in each Conv-Layers is not changed.
A MaxPooling Layer initialized in the init-function with Kernel of 2 and a stride of 2 bisects the image width and height every time when it occurs in the forward pass. It is used for feature reduction.
In the forward pass I start with a Conv layer which is defined in the init function, apply a ReLU as an activation function and bisect the width and height of the image via a MaxPoling layer. This layer combination Conv-ReLU-MaxPool appears three times for a deeply enough feature extraction and an appropriate image size/feature reduction.
Next I apply a dropout regularization to reduce overfitting.
A first fully connected layer takes in all the 50176 parameters. This number is the result of the product of the image size of the last MaxPooling-Layer (28x28) with the final depth of 64. I set the number of out_features arbitrary to 500.
Before implementing the classifier I set another dropout layer to minimize overfitting.
In a second fully connected layer (classifier) I turn these 500 in_features into the number of final classes (number of dog breeds) which is 133.
As a final result the prediction (x) will be returned from the forward function.
(IMPLEMENTATION) Specify Loss Function and Optimizer¶
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_scratch, and the optimizer as optimizer_scratch below.
import torch.optim as optim
### TODO: select loss function
criterion_scratch = nn.CrossEntropyLoss()
### TODO: select optimizer
optimizer_scratch = optim.SGD(model_scratch.parameters(), lr = 0.01)
(IMPLEMENTATION) Train and Validate the Model¶
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_scratch.pt'.
!pip show pillow
#from workspace_utils import active_session
def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
"""returns trained model"""
# initialize tracker for minimum validation loss
valid_loss_min = np.Inf
# exist save-file, load save file
#if os.path.exists(save_path):
# print("load previous saved model ...")
# model.load_state_dict(torch.load(save_path))
#with active_session():
for epoch in range(1, n_epochs+1):
# initialize variables to monitor training and validation loss
train_loss = 0.0
valid_loss = 0.0
###################
# train the model #
###################
model.train()
for batch_idx, (data, target) in enumerate(loaders['train']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
## find the loss and update the model parameters accordingly
## record the average training loss, using something like
## train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
# clear the gradients of all optimized variables
optimizer.zero_grad()
# forward pass: compute predicted outputs by passing inputs to the model
output = model(data)
# calculate the batch loss
loss = criterion(output, target)
# backward pass: compute gradient of the loss with respect to model parameters
loss.backward()
# perform a single optimization step (parameter update)
optimizer.step()
# update training loss
train_loss += loss.item()*data.size(0)
#train_loss_2 = train_loss_2 + ((1 / (batch_idx + 1)) * (loss.data - train_loss_2))
######################
# validate the model #
######################
model.eval()
for batch_idx, (data, target) in enumerate(loaders['valid']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
## update the average validation loss
# forward pass: compute predicted outputs by passing inputs to the model
output = model(data)
# calculate the batch loss
loss = criterion(output, target)
# update average validation loss
valid_loss += loss.item()*data.size(0)
# calculate average losses
train_loss = train_loss/len(train_loader.dataset)
valid_loss = valid_loss/len(valid_loader.dataset)
# print training/validation statistics
print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
epoch, train_loss, valid_loss))
## TODO: save the model if validation loss has decreased
# save model if validation loss has decreased
if valid_loss <= valid_loss_min:
print('Validation loss decreased ({:.6f} --> {:.6f}). Saving model ...'.format(
valid_loss_min,
valid_loss))
torch.save(model.state_dict(), save_path)
valid_loss_min = valid_loss
# return trained model
return model
# train the model
model_scratch = train(50, loaders_scratch, model_scratch, optimizer_scratch,
criterion_scratch, use_cuda, 'model_scratch.pt')
(IMPLEMENTATION) Test the Model¶
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 10%.
# load the model that got the best validation accuracy
model_scratch.load_state_dict(torch.load('model_scratch.pt'))
def test(loaders, model, criterion, use_cuda):
# monitor test loss and accuracy
test_loss = 0.
correct = 0.
total = 0.
model.eval()
for batch_idx, (data, target) in enumerate(loaders['test']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
# forward pass: compute predicted outputs by passing inputs to the model
output = model(data)
# calculate the loss
loss = criterion(output, target)
# update average test loss
test_loss = test_loss + ((1 / (batch_idx + 1)) * (loss.data - test_loss))
# convert output probabilities to predicted class
pred = output.data.max(1, keepdim=True)[1]
# compare predictions to true label
correct += np.sum(np.squeeze(pred.eq(target.data.view_as(pred))).cpu().numpy())
total += data.size(0)
print('Test Loss: {:.6f}\n'.format(test_loss))
print('\nTest Accuracy: %2d%% (%2d/%2d)' % (
100. * correct / total, correct, total))
# call test function
test(loaders_scratch, model_scratch, criterion_scratch, use_cuda)
Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)¶
You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.
(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset¶
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively).
If you like, you are welcome to use the same data loaders from the previous step, when you created a CNN from scratch.
## TODO: Specify data loaders
loaders_transfer = loaders_scratch
(IMPLEMENTATION) Model Architecture¶
Use transfer learning to create a CNN to classify dog breed. Use the code cell below, and save your initialized model as the variable model_transfer.
import torchvision.models as models
import torch.nn as nn
## TODO: Specify model architecture
model_transfer = models.vgg19(pretrained=True)
if use_cuda:
model_transfer = model_transfer.cuda()
print('GPU is used')
print(model_transfer)
# Freeze parameters so we don't backprop through them
for param in model_transfer.parameters():
param.requires_grad = False
model_transfer.classifier[6] = nn.Linear(4096, 133, bias=True)
#model_transfer.classifier = classifier
print(model_transfer)
Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem.
Answer: I just replaced the last layer of the classifier of VGG19 pretrained model, i.e. classifier(6.)
This is important to adjust the number of classes. The number of classes in the VGG19 approach is 1000. In the dog-breed-project I only have 133 classes. Hence, only the number of 'out_features' in classifier(6) is replaced by 133.
(IMPLEMENTATION) Specify Loss Function and Optimizer¶
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_transfer, and the optimizer as optimizer_transfer below.
#criterion_transfer = nn.NLLLoss()
criterion_transfer = nn.CrossEntropyLoss()
optimizer_transfer = optim.SGD(filter(lambda p: p.requires_grad,model_transfer.parameters()), lr=0.01)
(IMPLEMENTATION) Train and Validate the Model¶
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_transfer.pt'.
# train the model
n_epochs=10
model_transfer = train(n_epochs, loaders_transfer, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'model_transfer.pt')
# load the model that got the best validation accuracy (uncomment the line below)
model_transfer.load_state_dict(torch.load('model_transfer.pt'))
(IMPLEMENTATION) Test the Model¶
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 60%.
# Try out the model on the test dataset of dog images
test(loaders_transfer, model_transfer, criterion_transfer, use_cuda)
(IMPLEMENTATION) Predict Dog Breed with the Model¶
Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan hound, etc) that is predicted by your model.
### TODO: Write a function that takes a path to an image as input
### and returns the dog breed that is predicted by the model.
# list of class names by index, i.e. a name can be accessed like class_names[0]
class_names = [item[4:].replace("_", " ") for item in train_data.classes]
def predict_breed_transfer(img_path):
# load the image and return the predicted breed
# transform iamege to tensor
image_as_tensor = transform_to_tensor(img_path)
# move model inputs to cuda, if GPU available
if use_cuda:
image_as_tensor = image_as_tensor.cuda()
# predicted output
output = model_transfer(image_as_tensor)
# convert output probabilities to predicted class
_, preds_as_tensor = torch.max(output, 1)
pred_index = np.squeeze(preds_as_tensor.numpy()) if not use_cuda else np.squeeze(preds_as_tensor.cpu().numpy())
return class_names[pred_index]
def display_image(img_path, title=''):
image = Image.open(img_path)
plt.title(title)
plt.imshow(image)
plt.show()
Step 5: Write your Algorithm¶
Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,
- if a dog is detected in the image, return the predicted breed.
- if a human is detected in the image, return the resembling dog breed.
- if neither is detected in the image, provide output that indicates an error.
You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and dog_detector functions developed above. You are required to use your CNN from Step 4 to predict dog breed.
Some sample output for our algorithm is provided below, but feel free to design your own user experience!
(IMPLEMENTATION) Write your Algorithm¶
### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.
def run_app(img_path):
## handle cases for a human face, dog, and neither
# human face?:
if (face_detector(img_path)):
print("You seem to be a human being")
predicted_breed = predict_breed_transfer(img_path)
display_image(img_path, title="Predicted breed: {}".format(predicted_breed) )
print("But you look like a ...")
print(predicted_breed)
# dog?:
elif dog_detector(img_path):
print("You seem to be a dog!")
predicted_breed = predict_breed_transfer(img_path)
display_image(img_path, title="Predicted breed: {}".format(predicted_breed) )
print("The dog breed is most likley ...")
print(predicted_breed)
# neither
else:
print("In this image does not seem to show a human face or a dog.")
display_image(img_path, title="...")
print("Try another!")
print("\n")
Step 6: Test Your Algorithm¶
In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?
(IMPLEMENTATION) Test Your Algorithm on Sample Images!¶
Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.
Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.
Answer: (Three possible points for improvement)
In my opinion the tool of transfer learning is very impressive. I only used 5 epochs for traini and got a really good accuracy for testing of about 88%
Ways to improve the result:
- There is still no overfitting after 10 epochs of training the classifier. I could enlarge the number of epochs to get even better results.
- I could add another Conv-MaxPool layer block to further increase the depth and hence the feature extraction.
- I could try to reduce the learning rate.
- I could use a different optimizer, e.g. Adagrad Optimizer with an adaptive Learning rate
## TODO: Execute your algorithm from Step 6 on
## at least 6 images on your computer.
## Feel free to use as many code cells as needed.
## suggested code, below
for file in np.hstack((human_files[:10], dog_files[:10])):
run_app(file)
